Rapidus Corporation announced a new suite of artificial intelligence design tools to support its Rapidus AI-Assisted Design Solution, known as Raads, positioning the technology as central to its Rapid and Unified Manufacturing Service concept. The company said that this technology approach will be rebranded to Rapidus AI-Agentic Design Solution, with multiple tools planned to be released starting in 2026 and provided to customers together with a process design kit and reference flows. The tools are aimed at tightening the link between design and manufacturing, with a focus on advanced process nodes.
The first tool in the suite, Raads Generator, is described as an electronic design automation tool based on large-scale language models. According to Rapidus, when designers input semiconductor specifications, Raads Generator outputs register-transfer level design data that is optimized for Rapidus’ 2 nm manufacturing process. By using large-scale language models, the tool is intended to translate high level design intent into RTL that is tuned for the company’s specific process capabilities, which could shorten early design cycles and improve manufacturability.
The second tool, Raads Predictor, targets the later stages of the digital design flow, providing RTL debugging and optimization for physical design, placement, and routing. Rapidus said that Raads Predictor will provide a power performance area estimate in a short period of time, giving designers earlier visibility into implementation tradeoffs. Designers can use Raads Generator to write design ideas and desired specifications as RTL source code and then input data into Raads Predictor along with Synopsys Design Constraints, allowing them to predict the power performance area of the silicon manufactured by Rapidus and iterate on design choices before committing to full physical design.
